Path to 2024 series highlights realities of American attitudes

Focusing on corporate political action, AI, immigration, and more, the Polarization Research Lab aims to dispel myths about partisan beliefs.

Wall with American flag to represent politically divided Americans.
Image: iStock/wildpixel

A lot of the work of the Polarization Research Lab—a collaboration between faculty and researchers at the University of Pennsylvania, Dartmouth College, and Stanford University—is to show the disconnect between perceptions of what others think and reality. The lab has been running a weekly tracking poll for nearly two years through the online public opinion research company YouGov.

“We’re walking around with this distorted image of what our group believes, what the other group believes, and also what the other group thinks we think,” says Yphtach Lelkes, associate professor at the Annenberg School for Communication and co-director of the lab.

For example, across 73,325 YouGov survey interviews, the lab found that Democrats think a much larger percentage of Republicans condone politically motivated violence than actually do, and vice versa.

“When people walk around with these misperceptions, it closes off any discussion,” Lelkes says. “You just think the other side is bad, and, when you think that, there’s no way to find common ground. There’s no way to work together.”

In its Path to 2024 series, the lab has issued a monthly report since February analyzing public attitudes toward a different issue, with a hope of dispelling myths about the beliefs of everyday Americans. Released reports cover views on partisan violence, immigration, artificial intelligence, corporate political action, and culture wars, showing trends not only by political affiliation but also by race, gender, age, and education.

Future reports will touch on the Supreme Court, the urban-rural divide in America, democratic backsliding, political science research on elections, and election integrity and denialism.

The report on corporate political action also showed common ground between parties. Only 23% of Republicans and 39% of Democrats agreed that companies should take public positions on social issues. They also shared cynicism about the intent of corporate activism: 75% of Republicans and 66% of Democrats agreed that companies engage in activism as a marketing tool.

The report on AI, misinformation, and democracy shows that a minority of both Republicans and Democrats think AI will make personal privacy, national security, and American elections better, and that independents are the most pessimistic in all three areas. Younger, more affluent, and non-white respondents reported being less worried about AI. Lelkes says his big takeaway from this report is that people are worried, but it’s unclear if they understand what AI is.

The Polarization Research Lab’s report on culture wars, based on 3,000 interviews conducted in May, found agreement on some issues—for example, only a minority of both parties support less restrictive gun laws—but a partisan split on critical race theory and removing books from public schools if parents find them inappropriate. The lab also found divisions in its report on immigration; 62% of Democratic respondents said they believe immigration is good for America, compared to 25% of Republicans.

“People’s attitudes on this stuff are not organically created. They’re not walking around with this notion or born with this idea that immigration is good or bad; they’re paying attention to what political elites are saying,” Lelkes says. He adds that politicians are not merely reflecting public opinion but polarizing the population, and polarization is a function of whether they have decided to stake a position on a given topic.

Partisanship means people will go along with their party, which reduces accountability, says Lelkes. He says that society needs citizens who think critically instead of holding attitudes like, “I don’t support this, but at least they’re stopping Democrats.”

“Partisanship is a clear divide in American politics, but it’s masking significant nuance and heterogeneity that exists. Once you see that there’s heterogeneity—within your group, even—it throws a wrench in the conformity wheel,” Lelkes says. “I think an important part of what we do is to give a real representation of what the public thinks, to break this red-blue bubble that people exist in and these stereotypes of one another and the other side.”

To stay up to date on future reports, sign up for the Polarization Research Lab’s newsletter at

The Polarization Research Lab is co-directed by Sean J. Westwood, associate professor of government at Dartmouth College; Yphtach Lelkes, associate professor of communication and political science at the University of Pennsylvania; and Shanto Iyengar, the William Robertson Coe Professor and Professor of Political Science and Communication at Stanford University.